“…In the audio domain the most widely used features are Mel Frequency Cepstral Coefficients (MFCC) [60], [61], [69], [70], [91], [97], [98], a spectral representation of sound that approximates the human auditory system's response. Other features include pitch [68], [84], [91], [97], [98], intensity [84], [91], [97], [98], Relative Spectral Perceptual Linear Predictive (RASTA-PLP) coefficients [60], [61], [91], Linear Predictive Coding (LPC) coefficients [60], [70], [91], harmonic to noise ratio [98], and formants [68]. It is common to include the first and second order temporal derivatives of features [60], [61], [91], [97], [98].…”